AI for Financial Advisors: Tools & Strategies for Growth

Mark Mersman, USA Financial - All right, welcome. I'm joined by Mr. Craig Kirkpatrick. Craig is the co-founder of Optimal AdvisorAI. I know this isn't your first stop in the industry, so I'm going to need you to give a little bit of a background. What were you up to before starting Optimal AdvisorAI and what kind of led you to start this crazy business?

Craig Kirkpatrick, Optimal AdvisorAI - I'm glad to be here. I've been in financial services for 30 years. Prior to founding Optimal, I was the owner of a mutual fund company. So I worked directly with financial advisors and financial services professionals for the last three years. A couple of years ago, we sold our firm. And at that time, I began using AI. This is two years ago, before it's been in the press like it is now. And I put a document in AI and it summarized it and gave me bullet points in like 20 seconds. And this was two years ago. I've done nothing else but that and just got very curious on how to work with these models. And now I'm a couple thousand hours in and 7,500 prompts later, we've launched a firm called Optimal and trained over a thousand financial professionals around the globe on how to use this technology.

Mark Mersman - What exactly are you trying to do with Optimal? Is there necessarily a product that you're putting out there or is it more education and resources?

Craig Kirkpatrick - Well, at Optimal Advisor AI, we simplify AI for financial professionals. So we show them how to use AI technologies in ways they never thought possible. We sort of open the mind's eye around what's achievable and show them actual use cases in their daily practices to save time, enhance connection, and drive growth. And as I mentioned, our perspective comes from the fact that we've trained over 1,000 financial professionals to use this tech to really to save between one to five hours a week. So we have courses, we do consulting, and we understand how to engage with these models, and actually look forward to sharing with you sort of what we've learned.

Mark Mersman - Let's start there. You said that you kind of started dabbling with AI in 2023-ish, somewhere in that neighborhood. What kind of changes have you seen just in that short timeframe? And what excites you the most about the business decision that you made to start Optimal because I think you've got a pretty wide open space ahead of you.

Craig Kirkpatrick - Well, the first thing is, the models have just gotten simply much better. But let me say this at the outset, because I think it's important and I come bearing really good news on this front - there's no technology or AI experience required to use this. This started out as a big curiosity for me. And I started teaching all my financial advisor friends some of the tricks that I've learned to consume more content or create better content. So what's happened literally in the last six months, the models and outputs have just gotten dramatically better. We'll talk probably about compliance and how do you work in the models in that regard. So the thing I'm most excited about is just the quality of the output that these models can deliver. And again, having literally been in the trenches with individuals on this for thousands of hours, seeing people who've never used this technology before engage with it, learn from it, and have a really safe time has been very cool.

Mark Mersman - And you and I met, you came and spoke at a conference that we hosted in the fall. I think a big part of what you do, and this is probably important for folks to understand, is that you're not necessarily selling a product or something that you've cobbled together with, you know, a chat bot or anything like that. Your goal is to really help them understand some of the core AI tools that are out there and how to leverage that in their practice. So, let's assume everybody listening knows nothing about AI, knows nothing about ChatGPT. Let's start at kind of the basics. What exactly is it and what are some of these models that are out there being used and what does that mean?

Craig Kirkpatrick - Let me step back on your first point there just to mention, and I think one of the reasons our methods have been so popular is because we focus on use cases. What are the five or six things you can do on a Monday to save time? And so while there's a lot of discussion about AI in general, we're really focusing on just the use cases around that. But for somebody that's never used AI before, let me just also say, you've been here before. Like if you think back when the first personal computer hit the scene, right? A lot of people didn't fully grasp their potential and they really had to learn how these machines could streamline tasks that once seemed daunting, right? Or I don't know, you remember the Bloomberg terminals, right? When they first came out, it revolutionized financial markets, but only for those people who invested the time to master it and learn it. And in the same way, that's what we're doing at Optimal. We host advisor courses. We have informational newsletters to help an advisor sort of unlock this power, help them learn how to engage with the model. So it can literally save you hours. I mentioned deepening client relationships. We can talk about use cases around how to connect with clients and ultimately helping it to fuel growth. But when you want to talk about practical use cases, right? If you've never engaged with these models, what are some of the things that you can do? And I'll mention two or three things that are at a high level. The first thing that AI is really good at is consuming content. So Mark, think about all the content that you consume on a daily, weekly, monthly basis; on markets, on clients, on planning. I mean, it's a lot. You can use AI to consume 10 to 20 times more information in less time, hopefully to make better decisions. So that's what we help teach people. A 200 page investment document or, you know, Warren Buffett's annual report that just came out and you want to get it summarized. Consuming content is a really great use case. The other great use case is just content creation. I think we're going to talk about that. How can you use it to create better emails, better newsletters, better LinkedIn posts? And then the one area that we focus on that's gotten a lot of attention here is just the business development side for an advisor. How can you use AI to be better prepared for a client meeting, for example, or help write a specific article targeted to a particular audience? So there's all these areas in anybody's day-to-day practice - looking at emails, writing reports, meeting with clients - where a simple use case with AI can really help you. And again, there's no technology or AI experience required. And the coding language is English.

Mark Mersman - Yeah, and I think really a big part of it for me, as I've become more reliant on it in my day to day, it's about not questioning the limits. And what I mean by that is like what I would normally go to Google for to do searches or even asking others on my team, there's a lot of times where you can just turn to some of these tools and you're going to get the answers a whole lot quicker. You can get them way more refined and polished. In my mind, a big part of your job and what you're trying to push out there is awareness and changing muscle memory. You know, for me, I've just had to start anytime I'm going to Google, wait, no, I'm not going to Google that. I'm going to reposition. I'm going in a different direction. And it's remarkable how much it's changed my day to day world.

Craig Kirkpatrick - You know, Mark, you're so right. One of the things from just a perspective standpoint, actually two big things. First, AI is just a new tool, a new technology. And secondly, there's this transformational change going on. Anytime this happens, like with computers or the internet, you need to learn a new skill. And so the new skill that everyone's going to learn is this thing called prompting or giving instructions. Because in many cases you can't treat it like Google. In fact, a lot of our advisor clients treat a model like ChatGPT as though they're hiring an assistant in their office, right? And this is an amazing assistant. It has access to everything that's ever been written. It can read a thousand page book in minutes. It can speak every language. It never gets tired. But normally when you hire an assistant, you have to train that assistant. This is exactly opposite. You're the one that needs to be trained. We're not used to giving instructions to a computer. And I think that's where we help out. How can you get better output, better answers? And again, Mark, just to honor you because I know you have leaned into this a bit and are sort of doing some ninja prompting with the models.

Mark Mersman - What's funny about that, and what I've found for me is that the better I get at asking and developing prompts, I treat it like you said, as a personal assistant, so much so that I find myself putting niceties in and saying please and thank you in the prompts. Like, here I'm telling the robot thanks, please, please do this. Just like you would do a human. Because you don't want to be a jerk, you want to be polite.

Craig Kirkpatrick - There was an article, somebody came out and said that responses are a little bit better if you're nicer with the AI. But I'm not sure about that. But again, you've used it, right? What do you think are one of your biggest challenges in engaging with this?

Mark Mersman - For me, it's more between my own ears and making sure that number one, the more detail I put into the prompt - and to be clear, I'm a fan of ChatGPT - but there's obviously others out there, and Craig, I'd love your take. You're going to be more knowledgeable on that side of it. But I think by and large, they're all kind of getting you down the same path. But I've found number one is being more detailed, and number two, when I don't get something that I like or I'm not happy with the output, just ask it again and you better figure out how to ask it a little bit more concisely or give more context for your prompt. And so it's really about just truly engaging with the tool, right? And the more you can give it, the more context - and we'll talk a little bit about prompt engineering; there are some basic frameworks that you can follow - but for me, I think the biggest obstacle first was just getting the muscle memory to go there to use it more. And then it was just pumping in as much as you can into that prompt to get the level of granularity and detail that you're looking for on the output.

Craig Kirkpatrick - So you've brought up such a key point, again, when we work with individuals. And a lot of people in the audience have engaged with the AI models, I'm sure. And they've gotten an output and it hasn't been great. It doesn't sound like me or it's too jargony. And they're like, wow, this AI thing isn't that great. But just to your point, being able to have the mindset, it's an assistant in my office. It's a coach, it's a partner. And if I don't like the output, I say I don't like the output. I just did a prompt this morning and I got an output from the model for an email. And I'm like, this sounds like a computer wrote it. Be more conversational. It was just a little tip and it came back with a better answer. You asked a question about how can your advisors who've never used it, what can they do? And again, most financial professionals we talk with are in this camp, right? They're too busy, they're skeptical, you know, all of that. If you're new to chat, or any of the models - and by the way, these are all public models - to your point, there's ChatGPT, there's Gemini, there's Copilot, there's Claude, there's a whole bunch of models. And your point was, you know, they're all generally doing a very similar thing. There's a big race going on and who's going to win? But as a user, what you need to do is learn the skill of prompting and giving instructions.

Mark Mersman - Well, what do you see is the biggest difference between the different models? I mean, you've probably experimented with them more than I have.

Craig Kirkpatrick - Listen, they all do have a particular area that they do pretty well. And there's different rankings on consuming content versus being analytical versus coding. And each one of them have their own little area. We've tended to sort of lean initially just on some of the open AI ChatGPT models day in day out for different use cases - for loading a 500 page document, for putting in a PDF and pulling out financial information into an Excel doc, all of that. ChatGPT has been, generally speaking, generating some consistent output, but they're all, so I don't know. We like that model the best. We like a search engine called Perplexity, which is like Google on steroids. But Copilot has a lot of security benefits around it for enterprises and so forth. I don't know if we want to talk about compliance on that. So, I think if you're brand new to this and you've never done anything, just go open an account on ChatGPT and ask it a question.

Mark Mersman - And so give me a sense for what do you think are some of the biggest misconceptions that financial advisors have about using AI in their practice? Is there anything that jumps out at you?

Craig Kirkpatrick - Actually, what's been interesting is that I think most advisors don't see a use case for it because they've never used it. So wait, you mean I can write a LinkedIn article on a small business owner and get it published and sent through compliance and sound like me? I can do that in two minutes? There's no awareness. Or a client calls in and says, what's the difference between a Roth IRA and a regular IRA? They've answered that question their whole life. They don't know they can just type that question in and they'll get a draft immediately. My biggest thing is for those that haven't used it, they don't realize there's 20 different things they could do each week to save time. So they're not aware of the use cases. There is a compliance aspect to this that we just should really touch on because it's keeping a lot of people from doing, but two things, just real quick. The two biggest concerns are one, you're going to put a client document into one of these AI models, right? That's what compliance is worried about. Okay. So first off, newsflash, don't do that. Like you wouldn't put a client statement in Google. So don't do it here. That's the first one they get wigged out about. The second one is that you're going to rely on information that isn't correct. So let's say, Mark, that you're going to write a LinkedIn article or publish a newsletter about a market update in 2025 and you get some information, then compliance is worried that it's not correct. So we teach people how you can cite your sources, how you can verify the information, how you can do that. So those are easy ways to engage with these models to lower the compliance angst around using.

Mark Mersman - Well, let's talk a little bit about the degree of concern that both an advisor and a compliance department should have. How often are you seeing incorrect information? And you mentioned the term hallucinations before, so let's at least unpack that a little bit. Touch on those things.

Craig Kirkpatrick - So, first thing is all of your compliance structures that you have in place apply here. Okay, so that's the first thing. But let's say you're going to write a letter to a client about having them do new business with you, for example. All of your compliance items are still there. And you want to make sure the email is not promissory and doesn't guarantee and all that. You can actually tell the AI to act like an SEC or FINRA compliance review person and it will write the email in such a way. However, let me mention one other thing. When we were at your conference, and you and I've had a bunch of conversations around this, there wasn't one client document piece of information that was used in these use cases, right? So the worry of compliance is correct, but when you look at how you're using it, you're not engaging with client financial data.

Mark Mersman - Yeah, and I think that's the big one is PII. When that term hallucinations is thrown around, define that from the point of AI.

Craig Kirkpatrick - Yeah, sorry. And by the way, when you said what I'm excited about from today versus six months ago, these models have a tendency to want to please and so they'll make stuff up. They literally will make things up and nine months ago, it was kind of bad. Maybe 30-40 percent of it was, you know.

Mark Mersman - Well, we were in the thick of political season at that point, so maybe it was just following suit with politicians.

Craig Kirkpatrick - There's a famous case where a lawyer submitted all his case studies to court and every case study was made up. This was probably 18 months ago. So that made the headlines and everybody worries about that. This is where, and today the hallucination factor is maybe 2-4 percent. It's not that bad. But one of the messages, and I want to make sure I get this out on this podcast, that we tell everybody is actually the older you are and the more life and business experience you have, you have a huge advantage in using AI because human judgment matters and the output that you get, you'll know from 20 years in the business, was that reasonable? Is that not reasonable? And so, by the way, I'm working with some young college kids getting their resume and going up and interviewing and these outputs will come from them. And they're like, that's great. They'll cut and paste and send it off. And I'm like, no, no, it's not right. So it's a first time in technology ever that the older you are, you have an advantage in using it. And we're just trying to get those folks, you know, 40 and older or whatever, to engage with things because you sort of have an advantage in using it.

Mark Mersman - Well, and I think the message out of that, and I've had a college professor friend of mine share the same thing that he's like, some of these students, they literally just do a full copy paste and it is what it is and just accept it as done. I think it's important to understand to use this as a time saving tool, but it isn't going to do 100 % of the work. In order to make it right, you're going to have to put a little time and effort into it. You know, it got you 90 % of the way there maybe, now take it home the right way.

Craig Kirkpatrick - It's not here to do your work for you. It's here to do your work with you. And so all of a sudden, you have access to this tool as an assistant, a partner. You want to engage a Boston consulting quality consultant? Well, just go to chat and tell it to act like a Boston consulting quality executive and answer my questions. Or, answer this question as a Nobel laureate. Let me mention the big thing on generative AI that that's got everybody talking really at a high level is we're just moving from searching to getting answers.

Mark Mersman - It's funny that you say that because there's a lot of talk in marketing circles right now that AI is going to kill search engine optimization. You know, there's all these agencies out there that, their business is SEO. Even Google's got their own generative AI tool that comes up when you do searches. And, in a weird way, they're neutering their own business when they've got that because all the output below that, people are getting the answer that they want first and they don't have to search anymore.

Craig Kirkpatrick - That's so spot on. And, what happens when a billion people move from searching to getting answers? What happens when your client, and maybe we're talking about the future of financial planning, but what happens when your client moves from searching to getting answers? How will financial planning change? What do you need to do to address those sort of issues? 

Mark Mersman - I want to touch on that in a second. First, let's start with practical applications. So  when you're working with advisors, and I'll even emphasize staff probably get as much or more leverage out of it, but what are some of the basics? Like what's the blocking and tackling type of things that you say you really should start looking at using it for this stuff?

Craig Kirkpatrick - Let's talk business development. You can use this to write emails better and faster and stronger and all that. That's easy. It's solved the blank sheet of paper problem, right? You know, it's so much easier to edit somebody else's work. So first on the communication side, what's the difference between a Roth IRA and a regular IRA? It will write me a client letter in 10 seconds and now I can look at it and see what I like. But let's talk about business development. This just happened last week. We've got a client who is about to meet with a home builder in Seattle. He's never met with the client. They own a home building company and this gentleman's coming in to the office. And so how can you use AI to better prepare for this client meeting? So we helped him craft a prompt, send the AI up to the homebuilders website. And by the way, the advisor really wasn't familiar with the homebuilding business. So can you give me 10 questions I could ask this homebuilder client to better engage with them in my first meeting? Or what should I know as a planner about the homebuilding business so my client thinks that I know about that? In seconds, you're now better prepared to ask the question about seasonal cash flow of your business, for example, that you might not have known. So just that alone on the business development side to prepare for a meeting is one use case.

Mark Mersman - We've seen kind of similar type of deal. An advisor is looking to host a client appreciation and you know, give me 20 ideas for things I could do with my client in a small group in Grand Rapids, Michigan. And there you go. I mean, a lot of times you're going to know some of them, especially if you've been been there, but I was pleasantly surprised at some of the suggestions that I've seen. I didn't realize that was there or that was an option.

Craig Kirkpatrick - Let's say you have a person in your office who plans your event. AI can give you the 10 local events, unique event ideas. It can then give you an agenda of the evening and talk about ways to engage your clients. If you're going to go to an art museum, what are the six ways we can engage? So it creates the agenda and then it creates the invite. This is done in under five minutes, which might take your meeting planner a couple hours.to do. So hosting client events is a big one. Let me mention one as just using AI as a strategic coach. We worked with an advisor, 150 million out of Irvine, and how can they double the size of their practice? Create for me a strategic business plan to help me grow my business. Again, I'm using AI as a strategic coach in that regard. Or creating LinkedIn contact. Now, a lot of advisors just don't post on LinkedIn. It's a hassle, it's compliance, and I don't know what I'm going to say, right? You can create six LinkedIn posts. Let's say you want to target small business owners. Give me five ideas for small business owners that I could post on LinkedIn. Tell me when the best time to post is. Write the post, make it compliant. And again, you could do all that in a tenth of the time.

Mark Mersman - And all of that can be done with just the basic models. It's funny because as you were talking there, for us, when we record this podcast and this video, we're using AI a number of different times. So beforehand, I told ChatGPT who you are, what the interview is going to be about, and to give me some questions and some ideas, right? Just to help me get some conversation starters. So that was before. Now, when we get done with this recording, I'll get an AI transcript with this video tool, and I'll take that AI transcript and what we'll do with it is a few things. Number one, I'm going to pop it in. It will generate social media posts. Hey, give me 10 social media posts from this and then give me a summary for the YouTube page and the podcast page based upon this entire transcript. We'll also do some video editing, so it will take little cuts of the video and serve up a handful of 30 second little Shorts, all done through AI without ever having to lift a finger. And you kind of go through and pick the few that seem to work. It's amazing. The amount of time that it has cut just from doing something like this, from planning and then the back end execution, multiple hours saved for every episode.

Craig Kirkpatrick - Just hearing you describe that sounded really complicated, but I know what you had to do. So to our listeners who've never engaged with AI - again, this is some of the stuff we do in our workshops and our courses -  something that you just described can be done pretty easily. So let's say you want to plan a trip to Italy. You know, I'm going to Italy for five days with my wife and daughter, help me plan this trip. And you go to Google and it's really difficult. But go to ChatGPT and it will give you an itinerary. We also said, well, wait, I know where my hotel is, but my wife is gluten free and my daughter's a vegan and I'm with my 80 year old mother, what do I do? And you can just instruct the AI to find my hotel, read every menu on every restaurant and give me gluten free and vegan in Italy in seconds. So, everybody travels and just seeing a use case like that, it's again, a technology that you can use in your everyday life.

Mark Mersman - Yeah, that's everyday life. Also, I know we've talked about this, that same scenario using it. You know, you had a client that comes in and they're going on that trip in a few months or whatever. We've done it where we've used it with somebody that had a certain type of surgery and asked, "Hey, we want to get a gift for this person. Just had this type of a surgery. They're 50 years old. What suggestions do you have and provide the links" and it popped up certain links to different gifts that were out there that would apply for that person. It legitimately takes a ton of the thinking off the table to come up with creative ideas and creative ways to serve your clients and differentiate yourself. Which is the goal of it. So what other ideas do you share in terms of practical applications time-saving wise for advisors in their offices? Anything else that you try to push?

Craig Kirkpatrick - We talked about creating personalized emails, hosting client events, and just a newsletter, for example. If you have a newsletter that you need to send out, using AI as a newsletter partner to write it, it's better content, and just saves time around that has been really helpful. think about, again, two areas: consuming content. You mentioned the video. Ray Dalio, who I follow and just was on a podcast for an hour and I didn't have time to listen to it. I plugged it into AI, give me the three summaries, give me the three bullets, give me three action items I could take. So, consuming content, any area where you want to look at a fund. We have a client who the client called the advisor and said, "My friend wants me to invest in a real estate deal and I want you to look at it. What's your advice?" And you normally might not want to do that, but with AI, you can look at that document. You can get a summary. You can even create talking points around compliance and the risks associated with this investment. So you talk about, I'm responding to my clients quicker. I'm giving them a better email that's more specialized to them. My content that I'm sending to them is more specific. Until you begin to engage with AI, you don't even know you can do it. But those are just a few of them.

Mark Mersman - Yeah, we've seen, and this is something that I've done a bit too here and there. We have some advisors that have YouTube channels and that sort of thing, and they're always looking for ideas on content. And before AI, you know, there was a lot of it out there, but you had to kind of process and synthesize it and summarize. Now, we'll see a lot of advisors, they find a couple of good videos from others out there and just pop the link into one of the GPTs and say, "Hey, summarize this and I'm going to do a video on it and give me the bullets for me to work from a video".

Craig Kirkpatrick - Yeah, sure. How about a couple of others? How about your company meetings? Are they good? Are they marginal? Can I improve my culture? What are four things I can do to just improve the vibe around this office? Or, I want to hire an administrative assistant. Give me an ad that I could place to hire a person. So as you begin to think about what do I do all day long, you're like, wait, I could use my assistant to help accelerate. And that's what we do at Optimal in terms of helping people engage with the models better so they can get a better output. Because you can try all this at home, and I think you should just to engage. With a couple changes to the prompt and a couple instructions, the output can be increased by 100%.

Mark Mersman - Let's talk a little about prompt engineering then. When you're giving guidance there, best practices for how to write a good prompt, how do you coach to that? What's kind of your message there?

Craig Kirkpatrick - We have four courses here where we teach this stuff. In the first course, we try to teach what are the some of the foundational aspects of this. One of the keys when you're instructing the model is to provide context. And what does that mean? Like here you have this assistant you've just hired, they're in your office, they're ready to help. But that assistant doesn't know who you are or are you a gardener or are a chemist? So I'm telling my assistant, wait, I'm a financial planner. I'm trying to help my clients. I need to respond to this question. I'm giving them context to help my assistant understand who I am. That simple little twist can improve your output.

Mark Mersman - And I would also say, within that context, who's the audience? Who's this really geared towards? And the other fun thing that you can do with some of these, you can tell it to act who you want the AI model to be. If you want the AI model to write that email in the tone of Snoop Dogg, you can do that. Or, like you were saying, Boston consulting group. A lot of times when I'm working on web copy I'll tell chat, "Hey, you are an expert SEO and website content writer, blah, blah". Or I'll pick some of my favorite marketing gurus out there that are well known and, you know, write it in this tone.

Craig Kirkpatrick - We teach, how can you get the AI to sound more like you, right? Because that's the big rub with new users here. So listen, Eric Schmidt, the ex CEO of Google, who's this global force in AI education, just says that this is the most transformational technology in the history of the world. Kind of like electricity. And so I think for those folks that haven't had a chance to engage with it, you can do this because maybe you were around before computers and the Internet and a cell phone and you figured that out. You can. And it takes a level of curiosity and a willingness to just want to engage. One other thing that we ask our clients early on is what do you love to do? Do you want to spend more time with your kids or you love golfing or whatever. Okay. Do you want to do more of that? Well, yeah. Well go learn AI and save one to two hours a day. And then go do more of that stuff.

Mark Mersman - You know, it's funny because I think we've seen at different points - you know, I've been in the industry now, call it 25 years - and we've seen different points just in that tenure for me where advisors have said, "This is going to do away with financial advisors, right? This is going to eliminate the need for financial advisors, right?" And it definitely feels like AI, that sentiment is coming in a lot of professions. What's your take on that? Like, what do you see as the future of the financial advice space?

Craig Kirkpatrick - Great question we get asked all the time. Let me first say we don't think it's humans versus AI. That's not what it is. It's humans with AI versus humans without. Or in this case, it's advisors with AI versus advisors without. But when we think about the future of financial planning, let's just say that it's really hard to know. There's a couple of things that jump out to us. First off, client's expectations that we think are going to shift. Right? There's going to be more personalized, more proactive advice that's going to be required because they're going to have a lot of answers that they used to go to you for. So I think it's more personalized. Human skills are going to be way more valuable. So emotional intelligence, it's always been important. But if you're a tax person and you just deliver facts to your clients, that's gonna be problematic. So I think the human skill side is gonna be really valuable. I think comprehensive financial planning is gonna differentiate advisors. Like the full picture, do more with data. I think that can give someone an advantage. And then continuous learning and adaptability, I think is gonna be critical for success. Those people that didn't adopt the Bloomberg terminal, they were at a disadvantage.

Mark Mersman - And it's going to be interesting to see. I had a conversation with a company that they are building AI salespeople. So basically it's a fake human voice sounds very real that has been trained on every objection known to man for that company. It listens to hours and hours of real people having sales calls. Now, I think it's going to be a while before that makes its way into our industry for all the regulatory and compliance issues with it. But what he was saying was, you know, we're not that far from five, ten years from now, people are going to prefer to actually have a conversation with an AI robot. And I think what he was getting at was in lieu of the phone tree and dialing for numbers and waiting, they would rather get their immediate gratification from an AI robot. I just had my first call on it. My HVAC company, I called to set an appointment and it was an AI robot that took all my information And I don't think the average person, not that I'm trying to say I'm above average, but I kind of sensed that it was happening, so I asked some questions a few different ways to get it. But it was pretty darn good. And candidly, that beat me sitting there waiting five minutes for somebody to get off another line.

Craig Kirkpatrick -  Listen, that market's going to be impacted hugely. But when I think about the financial advisor and the relationship with their client. Like that trust is completely golden. And I tell you what, when some child doesn't get into a college or someone passes away, they're going to want a human connection. This was why I said, if you're an advisor, think about your clients. How much emotional connection do you have with them? Because that is what's going to cause people to want to work with you. When I want to go talk to a human and connect on life/emotional stuff, I think that's going to make that way more valuable.

Mark Mersman - And I think people want to work with somebody that they feel gets them, that they can relate to. And I've always joked, even when the robo-advisor thing came a number of years ago and advisors were concerned, "Oh, this is going to be the death of the financial advisor". You know, a robot can't give you a hug. The robot's not going to laugh with you and celebrate with you in that human way. I think the message and the takeaway is, start using this now so that you're saving time and you're reallocating some of that time towards the human component, the emotional component, right?

Craig Kirkpatrick - And by the way, if you don't know to do that, ask chat, what are 10 ways you can help me connect with my client? That's where this is going, Mark. And those advisors who work more on life coaching or empathy or connection will have a huge advantage. And they'll use AI to save time. Where now I have four more hours a week. What if you had four hours a week to call that client to take them out to lunch, to connect instead of doing, you know, portfolio or whatever you're doing that an AI robot could help you with an hour. So that's the twist. That's where it's going.

Mark Mersman - You know, if the writing hasn't been on the wall before, it is now. Your value as a financial advisor, if you're putting it all on the performance of a portfolio or the performance of whatever it is, you're going to be missing the boat because we're going to start to see AI things that come up that are going to look really impressive performance wise, I'm sure. And at the end of the day, if that is the drum that you're trying to beat for why a client should stay with you, I think you're going to live by that and die by that.

Craig Kirkpatrick - If you can leverage, if your office, if every person can save one to five hours a week, what are they freed up to do? More business, more growth, more whatever. I think there becomes a competitive advantage that a group can have. So if you think of chat as an assistant that is going to come into your office and change your workflow, how much time would you spend with that? What would you pay to bring in an assistant to do all this sort of stuff? Thinking of it as an assistant in your office is an interesting mindset because it sort of wraps your head around how much time do I want to spend learning this new thing, and then, how much do I want to spend to get educated here? That's the question.

Mark Mersman - So, as we wrap up here, I'll fire a couple of things at you. What's been your favorite success story that you've seen or heard from advisors that you work with where somebody took the plunge and, it might be just a crazy wild anecdotal story, but has anything come to mind?

Craig Kirkpatrick - So two weeks ago, an advisor heard that their client was going to Greece. And this is a client that's always grumpy to the advisor and never says anything. And he found out they were going to Greece and here's where they were going. And the rep got on ChatGPT and found three cool events within their hotel that they could do. And he just sent an email, "Hey, I know you're going to Greece" and gave them three ideas. And this took five minutes. And he just called me on this and he's like, this woman called me back and she's never called me and thanked me for like these great ideas or whatever. Anyway, maybe that seems small, but that didn't take very long. And then I think the other kick that we get is just seeing people get to save time. We've got this big NPS score for our courses and we get all these rave reviews and all that sort of stuff. Listen, I'm the son of two school teachers and to be able to get to teach about this product and see people engage with it in a way that can help save them time.

Mark Mersman - But you know what, the first story that you shared, I think it speaks to the human part of it. And it speaks in my mind that the work that people really appreciate from a financial advisor is oftentimes not financial at all. And I think once we realize that yeah, this might be the job that I'm doing for you with the numbers and the moving the accounts around, but the reality is that the real work that is truly valued probably isn't the portfolio stuff as much as you might think. So, what mistakes have you seen or pitfalls would you try to tell advisors to avoid?

Craig Kirkpatrick - I think they'll use it a little bit, but they won't make it a habit. They learn about it, and then they're excited, and then they kind of forget. So we really tell people, make 15 minutes a day just to begin to kind of use this. It's the adoption side. We do a good job at changing human behavior, but they'll use it. But then here's what happens. Because a lot of our advisors are really successful, they get back into their old ways of doing things, and they don't know that AI can help solve that. Listen, this is a new tech thing. It's hard to change human behavior. And I see people sort of struggle with that a little bit on the adoption side.

Mark Mersman - Okay. Any final parting shots on first steps an advisor should take, or a simple AI use case to try to immediately see some value, anything like that?

Craig Kirkpatrick - Maybe I could ask chat, hey, there's a mobile app here. "Hey chat, I'm a brand new financial advisor. I've never used chat before. Can you give me two simple steps to be able to engage with you?" "Absolutely. Step one, just start by asking any question or sharing what you'd like help with. Step two, I'll respond with the info or guidance you need. It's that simple." Anyway, there's an app for that. I still like the desktop application. But this is Siri on steroids. It didn't know I was going to ask that. And it knew I was a financial planner and gave me two ideas in seconds.

Mark Mersman - So, somebody wants to explore what you're doing more, how do they find out? You know, I got to give you a chance to plug what you're doing there, Craig.

Craig Kirkpatrick - Listen, we speak nationally on this topic. We're Optimal AdvisorAI. We have monthly courses that are 75 minutes for advisors that they can join and you sit through that course and you're going to learn an introductory 'how to use AI'. We have four courses that people can be part of. We also have a monthly newsletter that comes in with videos and tips on how to do AI. We do speaking around the country and we're just out educating on what are the use cases? So there's no AI or tech experience required here. Everything that I've talked about today, you can learn. So what our courses do is accelerate your interest in using AI. 

Mark Mersman - Your website is OptimalAdvisorAI.com. Well, hey, Craig, this is awesome. I feel like I can have this conversation ad nauseum. I mean, I'll probably have to have you back on at some point, especially the pace at which this stuff is changing.

Craig Kirkpatrick - I'd love to, Mark. And again, I honor you for engaging with AI, even though you have a very busy life. That's great.

Mark Mersman - It's the cheat code, right? It's really what it is. So, hey, I appreciate it a ton, Craig. And we will catch up soon. And thanks for joining us on the podcast.

Craig Kirkpatrick - Thanks, everyone.

-- 

The Financial Advisor Marketing Playbook is a podcast/video series for high-performing financial planning professionals that are committed to improving their craft, helping their clients, and growing their business. Hosted by Mark Mersman, Chief Marketing Officer at USA Financial, this series contains a wide variety of content – from quick win ideas to long-form interviews, each episode provides actionable marketing ideas and insights that can be implemented easily into your practice. From digital marketing to traditional direct-response marketing, each episode delivers straight-forward and engaging content that any financial professional can use to improve their bottom line and grow their practice.

Financial Advisor Marketing Playbook is also a podcast! Subscribe today via Apple Podcasts or your preferred podcast listening service for easier on-the-go listening.